bryanck opened a new pull request, #12613: URL: https://github.com/apache/iceberg/pull/12613
Spark loads jars from the classpath into the app classloader. The Spark submit command loads jars specified in the command into a classloader that is a child of the app classloader. If the Iceberg jars are loaded via the classpath, then any catalog, file IO, or AWS factory implementation classes in the user jar cannot be loaded. This PR adds an option to the dynamic constructor utility, to enable falling back to the current thread context's classloader. It also enables the new option when loading catalog classes, file IO classes, and AWS factories. It seems fairly safe to always fallback to the Thread context's classloader, but I added an option to be on the conservative side. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For additional commands, e-mail: issues-h...@iceberg.apache.org
